cv::cuda::HostMem 类参考

一个类，它封装了CUDA的特殊内存类型分配函数，具有引用计数。有关详细信息，请参阅这里

#include <opencv2/core/cuda.hpp>

cv::cuda::HostMem 的协作图

公共类型
枚举	AllocType {   PAGE_LOCKED = 1 ,   SHARED = 2 ,   WRITE_COMBINED = 4 }

公共成员函数
	HostMem (const HostMem &m)
	HostMem (HostMem::AllocType alloc_type=HostMem::AllocType::PAGE_LOCKED)
	HostMem (InputArray arr, HostMem::AllocType alloc_type=HostMem::AllocType::PAGE_LOCKED)
	从主机内存复制数据创建
	HostMem (int rows, int cols, int type, HostMem::AllocType alloc_type=HostMem::AllocType::PAGE_LOCKED)
	HostMem (Size size, int type, HostMem::AllocType alloc_type=HostMem::AllocType::PAGE_LOCKED)
	~HostMem ()
int	channels () const
HostMem	clone () const
	返回矩阵的深拷贝，即数据被复制
void	create (int rows, int cols, int type)
	除非矩阵已具有指定的尺寸和类型，否则分配新的矩阵数据。
void	create (Size size, int type)
GpuMat	createGpuMatHeader () const
	将CPU内存映射到GPU地址空间，并创建不带引用计数的cuda::GpuMat头。
Mat	createMatHeader () const
	返回不带引用计数的HostMem数据矩阵头。
int	depth () const
size_t	elemSize () const
size_t	elemSize1 () const
bool	empty () const
bool	isContinuous () const
HostMem &	operator= (const HostMem &m)
void	release ()
	减少引用计数并在需要时释放内存。
HostMem	reshape (int cn, int rows=0) const
Size	size () const
size_t	step1 () const
void	swap (HostMem &b)
	与其他智能指针交换。
int	type () const

静态公共成员函数
static MatAllocator *	getAllocator (HostMem::AllocType alloc_type=HostMem::AllocType::PAGE_LOCKED)

公共属性
AllocType	alloc_type
int	cols
uchar *	data
const uchar *	dataend
uchar *	datastart
int	flags
int *	refcount
int	rows
size_t	step

详细信息

带有CUDA特殊内存类型分配函数引用计数的类。

其接口类似于Mat，但增加了内存类型参数。

• PAGE_LOCKED 设置页面锁定内存类型，通常用于从/到GPU的快速异步上传/下载数据。
• SHARED 指定零拷贝内存分配，如果支持，可以将主内存映射到GPU地址空间。
• WRITE_COMBINED 设置写合并缓冲区，不会被CPU缓存。此类缓冲区用于向GPU提供数据时，GPU只读取数据。其优点是优化CPU缓存使用。

注意

此类内存的分配大小通常有限。更多详细信息，请参阅《CUDA 2.2 Pinned Memory APIs》文档或《CUDA C Programming Guide》。

成员枚举文档

AllocType

枚举 cv::cuda::HostMem::AllocType

枚举值
PAGE_LOCKED
SHARED
WRITE_COMBINED

构造函数 & 析构函数文档

HostMem() [1/5]

cv::cuda::HostMem::HostMem ( HostMem::AllocType  alloc_type = HostMem::AllocType::PAGE_LOCKED )

explicit

Python
	cv.cuda.HostMem(	[, alloc_type]	) ->	<cuda_HostMem object>
	cv.cuda.HostMem(	rows, cols, type[, alloc_type]	) ->	<cuda_HostMem object>
	cv.cuda.HostMem(	size, type[, alloc_type]	) ->	<cuda_HostMem object>
	cv.cuda.HostMem(	arr[, alloc_type]	) ->	<cuda_HostMem object>

HostMem() [2/5]

cv::cuda::HostMem::HostMem

(

const HostMem &

m

)

Python
	cv.cuda.HostMem(	[, alloc_type]	) ->	<cuda_HostMem object>
	cv.cuda.HostMem(	rows, cols, type[, alloc_type]	) ->	<cuda_HostMem object>
	cv.cuda.HostMem(	size, type[, alloc_type]	) ->	<cuda_HostMem object>
	cv.cuda.HostMem(	arr[, alloc_type]	) ->	<cuda_HostMem object>

HostMem() [3/5]

cv::cuda::HostMem::HostMem	(	int	rows,
		int	cols,
		int	type,
		HostMem::AllocType	alloc_type = HostMem::AllocType::PAGE_LOCKED
	)

Python
	cv.cuda.HostMem(	[, alloc_type]	) ->	<cuda_HostMem object>
	cv.cuda.HostMem(	rows, cols, type[, alloc_type]	) ->	<cuda_HostMem object>
	cv.cuda.HostMem(	size, type[, alloc_type]	) ->	<cuda_HostMem object>
	cv.cuda.HostMem(	arr[, alloc_type]	) ->	<cuda_HostMem object>

HostMem() [4/5]

cv::cuda::HostMem::HostMem	(	Size	size,
		int	type,
		HostMem::AllocType	alloc_type = HostMem::AllocType::PAGE_LOCKED
	)

Python
	cv.cuda.HostMem(	[, alloc_type]	) ->	<cuda_HostMem object>
	cv.cuda.HostMem(	rows, cols, type[, alloc_type]	) ->	<cuda_HostMem object>
	cv.cuda.HostMem(	size, type[, alloc_type]	) ->	<cuda_HostMem object>
	cv.cuda.HostMem(	arr[, alloc_type]	) ->	<cuda_HostMem object>

HostMem() [5/5]

cv::cuda::HostMem::HostMem ( InputArray  arr, HostMem::AllocType  alloc_type = HostMem::AllocType::PAGE_LOCKED  )

explicit

Python
	cv.cuda.HostMem(	[, alloc_type]	) ->	<cuda_HostMem object>
	cv.cuda.HostMem(	rows, cols, type[, alloc_type]	) ->	<cuda_HostMem object>
	cv.cuda.HostMem(	size, type[, alloc_type]	) ->	<cuda_HostMem object>
	cv.cuda.HostMem(	arr[, alloc_type]	) ->	<cuda_HostMem object>

从主机内存复制数据创建

~HostMem()

cv::cuda::HostMem::~HostMem

(

)

成员函数文档

channels()

int cv::cuda::HostMem::channels

(

)

const

Python
	cv.cuda.HostMem.channels(		) ->	retval

clone()

HostMem cv::cuda::HostMem::clone

(

)

const

Python
	cv.cuda.HostMem.clone(		) ->	retval

返回矩阵的深拷贝，即数据被复制

create() [1/2]

void cv::cuda::HostMem::create	(	int	rows,
		int	cols,
		int	type
	)

Python
	cv.cuda.HostMem.create(	rows, cols, type	) ->	None

除非矩阵已具有指定的尺寸和类型，否则分配新的矩阵数据。

create() [2/2]

void cv::cuda::HostMem::create	(	Size	size,
		int	type
	)

Python
	cv.cuda.HostMem.create(	rows, cols, type	) ->	None

createGpuMatHeader()

GpuMat cv::cuda::HostMem::createGpuMatHeader

(

)

const

将CPU内存映射到GPU地址空间，并创建不带引用计数的cuda::GpuMat头。

This can be done only if memory was allocated with the SHARED flag and if it is supported by the hardware. Laptops often share video and CPU memory, so address spaces can be mapped, which eliminates an extra copy.

createMatHeader()

Mat cv::cuda::HostMem::createMatHeader

(

)

const

Python
	cv.cuda.HostMem.createMatHeader(		) ->	retval

返回不带引用计数的HostMem数据矩阵头。

depth()

int cv::cuda::HostMem::depth

(

)

const

Python
	cv.cuda.HostMem.depth(		) ->	retval

elemSize()

size_t cv::cuda::HostMem::elemSize

(

)

const

Python
	cv.cuda.HostMem.elemSize(		) ->	retval

elemSize1()

size_t cv::cuda::HostMem::elemSize1

(

)

const

Python
	cv.cuda.HostMem.elemSize1(		) ->	retval

empty()

bool cv::cuda::HostMem::empty

(

)

const

Python
	cv.cuda.HostMem.empty(		) ->	retval

getAllocator()

静态 MatAllocator * cv::cuda::HostMem::getAllocator ( HostMem::AllocType  alloc_type = HostMem::AllocType::PAGE_LOCKED )

静态

isContinuous()

bool cv::cuda::HostMem::isContinuous

(

)

const

Python
	cv.cuda.HostMem.isContinuous(		) ->	retval

operator=()

HostMem & cv::cuda::HostMem::operator=

(

const HostMem &

m

)

release()

void cv::cuda::HostMem::release

(

)

减少引用计数并在需要时释放内存。

reshape()

HostMem cv::cuda::HostMem::reshape	(	int	cn,
		int	rows = 0
	)		const

Python
	cv.cuda.HostMem.reshape(	cn[, rows]	) ->	retval

为相同数据创建具有不同通道数和/或不同行数的不同HostMem头

size()

Size cv::cuda::HostMem::size

(

)

const

Python
	cv.cuda.HostMem.size(		) ->	retval

step1()

size_t cv::cuda::HostMem::step1

(

)

const

Python
	cv.cuda.HostMem.step1(		) ->	retval

swap()

void cv::cuda::HostMem::swap

(

HostMem &

b

)

Python
	cv.cuda.HostMem.swap(	b	) ->	None

与其他智能指针交换。

type()

int cv::cuda::HostMem::type

(

)

const

Python
	cv.cuda.HostMem.type(		) ->	retval

成员变量文档

alloc_type

AllocType cv::cuda::HostMem::alloc_type

cols

int cv::cuda::HostMem::cols

data

uchar* cv::cuda::HostMem::data

dataend

const uchar* cv::cuda::HostMem::dataend

datastart

uchar* cv::cuda::HostMem::datastart

flags

int cv::cuda::HostMem::flags

refcount

int* cv::cuda::HostMem::refcount

rows

int cv::cuda::HostMem::rows

step

size_t cv::cuda::HostMem::step

---

本类的文档是由以下文件生成的

• opencv2/core/cuda.hpp